Corridor-level signalized intersection safety analysis in Shanghai, China using Bayesian hierarchical models.

Area of Science:

• Transportation Engineering
• Urban Planning
• Traffic Safety Research

Background:

• Traffic crashes in Chinese cities predominantly occur at signalized intersections.
• Intersection safety research in China is nascent, necessitating advanced analytical approaches.
• Shanghai's dense road network features closely spaced signalized intersections (avg. 200m) with coordinated signals, implying inter-corridor dependencies.

Purpose of the Study:

• To analyze intersection safety in Shanghai by considering correlations within traffic corridors.
• To identify key risk factors contributing to traffic crashes at signalized intersections and along corridors.

Main Methods:

• Utilized data from 195 signalized intersections across 22 urban corridors in Shanghai.
• Collected mean speeds and speed variances using GPS data from taxis.
• Applied Bayesian hierarchical models to assess intersection and corridor-level crash risk factors.

Main Results:

• Intersections on corridors with lower mean speeds exhibited fewer crashes compared to those with higher speeds.
• Higher crash frequencies were observed at intersections situated along two-way roads, beneath elevated roads, and in close proximity to each other.

Conclusions:

• Corridor characteristics, including mean speed and spatial configuration, significantly influence intersection safety.
• Understanding and mitigating risks associated with specific intersection types and corridor attributes is crucial for improving urban traffic safety in Shanghai.